A/B Test Significance Calculator
Find out if your test results are statistically significant
How the A/B Test Calculator works
Enter your variants
Input visitors and conversions for both Variant A (control) and Variant B (challenger).
Two-proportion z-test
We run a proper statistical z-test to compare the two conversion rates and calculate a p-value.
Choose confidence
Pick 90%, 95%, or 99% confidence. Higher confidence means you need more data to declare a winner.
Get your verdict
See if you have a statistically significant winner, or how much more traffic you need.
Rate A
Control
Rate B
Challenger
Uplift
Confidence
What are you testing?
A
Variant A (Control)
B
Variant B (Challenger)
Confidence Level
Higher confidence requires more data but reduces false positives.
Variant converts
than Variant .
A
Control
B
Challenger
Winner:
More Traffic Needed
~ more visitors per variant
Estimated additional sample size needed to reach confidence with 80% statistical power.
Recommendations
Statistical significance does not guarantee real-world impact. Always consider practical significance and run tests for at least 1-2 full business cycles.
Optimize your tests
AI-generated ad
variations win more tests
MarketDragon creates high-converting ad creatives to fuel your A/B testing pipeline.